Wim Saeys
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Wim Saeys has a total of 135 co-authors. Here are the top co-authors by number of shared publications:
- Steven Truijen
Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, University of Antwerp, Universiteitsplein 1, 2610 Antwerpen, Belgium
27 shared publications Steven Truijen
Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp - Belgium
14 shared publications- Luc Vereeck
Department of Rehabilitation Sciences and Physiotherapy/Movant, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium
13 shared publications - Ann Hallemans
Research Group MOVANT, Department of Rehabilitation Sciences and Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium
11 shared publications - Tamaya Van Criekinge
Department of Rehabilitation Sciences, KU Leuven, Brugge, Belgium
11 shared publications Christophe Lafosse
Department of Neurorehabilitation, RevArte Rehabilitation Hospital, Edegem, Belgium
6 shared publications
Latest Publications
- Open Access
From Neurophysiological Mechanisms to Rehabilitation After Botulinum Toxin Type A in Post-Stroke Spasticity
Background: Botulinum toxin type A (BoNT-A) is globally recognized as a standard treatment for post-stroke spasticity (PSS), although its secondary mechanisms remain underinvestigated. The objective of this study was to investigate the neurophysiological mechanisms underlying BoNT-A treatment in PSS and explore their implications for mechanism-based rehabilitation. Methods: A systematic review and meta-analysis (PROSPERO registration number ID: CRD420261352230) of 39 studies involving 760 patients was conducted on 18 August 2026, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed, Web of Science, Scopus, and Embase were systematically searched using the PICO framework. Where possible, quantitative data were pooled according to predefined neurophysiological constructs, including passive stretch-evoked activity, voluntary activation of the injected muscle, compound muscle action potential (CMAP), reciprocal motor control, normalized Hmax/Mmax, and specific spinal inhibitory mechanisms. Results: Quantitative synthesis demonstrated construct-specific neurophysiological changes following BoNT-A treatment. Passive stretch-evoked activity showed the largest pooled effect (Hedges’ g = 0.80, 95% CI 0.35–1.24), followed by CMAP amplitude (g = 0.69, 95% CI 0.23–1.14) and reciprocal motor control (g = 0.52, 95% CI 0.23–0.81). Effects on voluntary activation of the injected muscle (g = 0.31, 95% CI −0.38 to 1.00) and normalized Hmax/Mmax (g = 0.22, 95% CI −0.74 to 1.17) were smaller and more heterogeneous. Reciprocal and recurrent inhibition were each represented by single-study estimates, while supraspinal outcomes were synthesized narratively due to methodological heterogeneity. Conclusions: The findings support a predominantly peripheral mechanism of BoNT-A while suggesting possible secondary changes within spinal and supraspinal motor control pathways. However, the certainty of evidence was low to very low. These neurophysiological effects provide a rationale for individualized, mechanism-based rehabilitation during the post-injection therapeutic window, although higher-quality evidence is required to confirm this framework.
Open AccessFrom Neurophysiological Mechanisms to Rehabilitation After Botulinum Toxin Type A in Post-Stroke Spasticity
Background: Botulinum toxin type A (BoNT-A) is globally recognized as a standard treatment for post-stroke spasticity (PSS), although its secondary mechanisms remain underinvestigated. The objective of this study was to investigate the neurophysiological mechanisms underlying BoNT-A treatment in PSS and explore their implications for mechanism-based rehabilitation. Methods: A systematic review and meta-analysis (PROSPERO registration number ID: CRD420261352230) of 39 studies involving 760 patients was conducted on 18 August 2026, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed, Web of Science, Scopus, and Embase were systematically searched using the PICO framework. Where possible, quantitative data were pooled according to predefined neurophysiological constructs, including passive stretch-evoked activity, voluntary activation of the injected muscle, compound muscle action potential (CMAP), reciprocal motor control, normalized Hmax/Mmax, and specific spinal inhibitory mechanisms. Results: Quantitative synthesis demonstrated construct-specific neurophysiological changes following BoNT-A treatment. Passive stretch-evoked activity showed the largest pooled effect (Hedges’ g = 0.80, 95% CI 0.35–1.24), followed by CMAP amplitude (g = 0.69, 95% CI 0.23–1.14) and reciprocal motor control (g = 0.52, 95% CI 0.23–0.81). Effects on voluntary activation of the injected muscle (g = 0.31, 95% CI −0.38 to 1.00) and normalized Hmax/Mmax (g = 0.22, 95% CI −0.74 to 1.17) were smaller and more heterogeneous. Reciprocal and recurrent inhibition were each represented by single-study estimates, while supraspinal outcomes were synthesized narratively due to methodological heterogeneity. Conclusions: The findings support a predominantly peripheral mechanism of BoNT-A while suggesting possible secondary changes within spinal and supraspinal motor control pathways. However, the certainty of evidence was low to very low. These neurophysiological effects provide a rationale for individualized, mechanism-based rehabilitation during the post-injection therapeutic window, although higher-quality evidence is required to confirm this framework.
- Open Access
- 30 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (11) , 4246
Neurophysiological and Structural–Mechanical Changes Associated with Dry Needling in Post-Stroke Spasticity: A Systematic Review
Background/Objectives: In the past few years increasing attention has been given to the application of dry needling (DN) for spasticity in stroke survivors. Nevertheless, the underlying mechanisms of this technique have not yet been confirmed. The aim of this systematic review was to distinguish the effects of DN in post-stroke spasticity on both structural–mechanical muscle properties (SMMPs) and neurophysiological properties to address these mechanisms. Methods: A literature search was performed in Web of Science, PubMed, Scopus and Embase following PRISMA guidelines (PROSPERO ID: 1163064). Randomized controlled trials and case–control studies involving adults with post-stroke spasticity treated with DN were included. Outcomes were categorized as SMMPs (e.g., muscle architecture, passive stiffness, PROM) or neurophysiological measures (e.g., H-reflex, H/M ratio). Standardized effect sizes (Hedges’ g) were calculated when possible; however, heterogeneity in outcomes and incomplete variance reporting precluded meta-analysis. Results: Twelve studies met the inclusion criteria. Most of these studies assessed passive range of motion, reporting a significant increase following the intervention. Only two of the included studies examined structural characteristics, and five studies included neurophysiological outcomes. Correlations between mechanistic outcomes and clinical spasticity grading (MAS/MMAS) were weak. Emerging evidence suggests DN may additionally modulate local inflammatory mediators, indicating a potential neuroimmune contribution to its effects. Conclusions: DN appears to improve structural–mechanical muscle properties and produce moderate reductions in reflex excitability in individuals with post-stroke spasticity. Mechanical adaptations are more consistently demonstrated than neural changes, and neither domain is proportionally reflected in clinical spasticity scales. Evidence remains limited by small samples, methodological variability, and incomplete reporting. Further mechanistic research is needed to clarify how DN influences the complex pathophysiology of post-stroke spasticity.
- 30 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (11) , 4246
Open AccessNeurophysiological and Structural–Mechanical Changes Associated with Dry Needling in Post-Stroke Spasticity: A Systematic Review
Background/Objectives: In the past few years increasing attention has been given to the application of dry needling (DN) for spasticity in stroke survivors. Nevertheless, the underlying mechanisms of this technique have not yet been confirmed. The aim of this systematic review was to distinguish the effects of DN in post-stroke spasticity on both structural–mechanical muscle properties (SMMPs) and neurophysiological properties to address these mechanisms. Methods: A literature search was performed in Web of Science, PubMed, Scopus and Embase following PRISMA guidelines (PROSPERO ID: 1163064). Randomized controlled trials and case–control studies involving adults with post-stroke spasticity treated with DN were included. Outcomes were categorized as SMMPs (e.g., muscle architecture, passive stiffness, PROM) or neurophysiological measures (e.g., H-reflex, H/M ratio). Standardized effect sizes (Hedges’ g) were calculated when possible; however, heterogeneity in outcomes and incomplete variance reporting precluded meta-analysis. Results: Twelve studies met the inclusion criteria. Most of these studies assessed passive range of motion, reporting a significant increase following the intervention. Only two of the included studies examined structural characteristics, and five studies included neurophysiological outcomes. Correlations between mechanistic outcomes and clinical spasticity grading (MAS/MMAS) were weak. Emerging evidence suggests DN may additionally modulate local inflammatory mediators, indicating a potential neuroimmune contribution to its effects. Conclusions: DN appears to improve structural–mechanical muscle properties and produce moderate reductions in reflex excitability in individuals with post-stroke spasticity. Mechanical adaptations are more consistently demonstrated than neural changes, and neither domain is proportionally reflected in clinical spasticity scales. Evidence remains limited by small samples, methodological variability, and incomplete reporting. Further mechanistic research is needed to clarify how DN influences the complex pathophysiology of post-stroke spasticity.
- Open Access
- 20 May 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (5) , e0334571
Spinal mechanisms and feasibility of Dry Needling versus Botulinum Toxin Type A in post-stroke lower limb spasticity: A proof-of-concept randomized clinical trial protocol (STROKE-POC)
Introduction: Stroke often causes spasticity, impacting mobility and quality of life. Botulinum Toxin type A (BTX-A) and Dry Needling (DN) are treatments that reduce spasticity, although Botulinum Toxin type A injections can cause adverse effects. No studies have directly compared their effects at spinal, muscular, functional, quality-of-life, and cost-effectiveness levels. This study aims to determine the spinal mechanisms of BTX-A and DN on post-stroke lower limb spasticity, while also assessing feasibility, safety, and exploratory effects at muscular, functional, quality-of-life, and cost-effectiveness levels. Methods and analysis: This is a protocol of a proof-of-concept, feasibility randomized clinical trial including 90 participants from Canada, Belgium, and Spain who experienced a first stroke in the previous 12 months and present plantar flexor spasticity. Time since stroke (0–12 months) will be recorded and explored as a potential modifier of treatment response. Participants will be randomly assigned to receive either one session of BTX-A or 12 weekly sessions of DN. Blinded evaluators will assess outcomes before, during, and after treatment, with a 4-week follow-up. The primary outcome will be spinal mechanisms of spasticity, measured using the Tonic Stretch Reflex Threshold and its velocity sensitivity. Secondary outcomes will assess: a) muscular architecture and echotexture (measured with ultrasound); b) muscle tone/resistance using the Modified Ashworth Scale; c) gait and mobility (instrumented analysis, Timed Up and Go, 10-Meter Walk Test); d) muscle strength with dynamometry; e) quality of life with the EuroQoL questionnaire; and f) cost-effectiveness (analytic model). The findings will provide preliminary data to inform a future definitive trial. Ethics and Dissemination: This research project has secured funding from the NEURON ERA-NET 2022 call, supported by the European Union’s Horizon 2020 program (GA 964215) and co-funded by the European Union-Next Generation, and has undergone peer review. Ethical approval has been obtained from Spain, Canada, and Belgium. The study is registered in ClinicalTrials.gov (NCT06296082) and the Clinical Trials Information System (CTIS) under the number 2024-510866-18-00. The study protocol is registered on Zenodo ( https://doi.org/10.5281/zenodo.20034064 ) Clinical Trials: Clinical Trials NCT06296082; https://clinicaltrials.gov/study/NCT06296082
- 20 May 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (5) , e0334571
Open AccessSpinal mechanisms and feasibility of Dry Needling versus Botulinum Toxin Type A in post-stroke lower limb spasticity: A proof-of-concept randomized clinical trial protocol (STROKE-POC)
Introduction: Stroke often causes spasticity, impacting mobility and quality of life. Botulinum Toxin type A (BTX-A) and Dry Needling (DN) are treatments that reduce spasticity, although Botulinum Toxin type A injections can cause adverse effects. No studies have directly compared their effects at spinal, muscular, functional, quality-of-life, and cost-effectiveness levels. This study aims to determine the spinal mechanisms of BTX-A and DN on post-stroke lower limb spasticity, while also assessing feasibility, safety, and exploratory effects at muscular, functional, quality-of-life, and cost-effectiveness levels. Methods and analysis: This is a protocol of a proof-of-concept, feasibility randomized clinical trial including 90 participants from Canada, Belgium, and Spain who experienced a first stroke in the previous 12 months and present plantar flexor spasticity. Time since stroke (0–12 months) will be recorded and explored as a potential modifier of treatment response. Participants will be randomly assigned to receive either one session of BTX-A or 12 weekly sessions of DN. Blinded evaluators will assess outcomes before, during, and after treatment, with a 4-week follow-up. The primary outcome will be spinal mechanisms of spasticity, measured using the Tonic Stretch Reflex Threshold and its velocity sensitivity. Secondary outcomes will assess: a) muscular architecture and echotexture (measured with ultrasound); b) muscle tone/resistance using the Modified Ashworth Scale; c) gait and mobility (instrumented analysis, Timed Up and Go, 10-Meter Walk Test); d) muscle strength with dynamometry; e) quality of life with the EuroQoL questionnaire; and f) cost-effectiveness (analytic model). The findings will provide preliminary data to inform a future definitive trial. Ethics and Dissemination: This research project has secured funding from the NEURON ERA-NET 2022 call, supported by the European Union’s Horizon 2020 program (GA 964215) and co-funded by the European Union-Next Generation, and has undergone peer review. Ethical approval has been obtained from Spain, Canada, and Belgium. The study is registered in ClinicalTrials.gov (NCT06296082) and the Clinical Trials Information System (CTIS) under the number 2024-510866-18-00. The study protocol is registered on Zenodo ( https://doi.org/10.5281/zenodo.20034064 ) Clinical Trials: Clinical Trials NCT06296082; https://clinicaltrials.gov/study/NCT06296082
- Open Access
- 29 January 2026
- Published by Aboutscience Srl in Archives of Physiotherapy
- Vol. 16 (1) , 1-10
Stair walking is associated with returning home after inpatient stroke rehabilitation in Belgium and Switzerland: a multicentric retrospective study
Introduction: Identifying factors associated with discharge destination after inpatient stroke rehabilitation is important for patients and healthcare professionals. It supports discharge planning and prevents delayed discharge.Objective: To identify key variables from socio-demographic and clinical data associated with returning home after inpatient stroke rehabilitation, focusing on patients from three rehabilitation centers in Belgium and Switzerland.Methods: This multicenter retrospective study, conducted in three centers, included 1475 adult patients with stroke admitted to an inpatient rehabilitation unit between December 2012 and June 2021. A logistic regression with backward selection was used to define the model for discharge destination. The dependent variable was the discharge destination (home vs other). The independent variables were selected from the socio-demographic, medical, neurological, care pathway, and functional data and included age, gender, living arrangement, type of stroke, previous stroke, cognitive impairments, independence in grooming, eating, and stair walking.Results: The final model included three variables (independence in stair walking, living arrangement, and cognitive impairment). Stair walking had the strongest association with returning home. Patients who were partially (OR 5.83, 95% CI 3.67-9.26) or fully independent (OR 14.31, 95% CI 9.34-21.93) were more likely to return home than patients who were unable to walk the stairs. The results were similar for subgroups and for discharge and admission data.Conclusion: The study showed that independence in walking stairs is strongly associated with discharge destination. Aligned with another study, these results should be confirmed in further research.
- 29 January 2026
- Published by Aboutscience Srl in Archives of Physiotherapy
- Vol. 16 (1) , 1-10
Open AccessStair walking is associated with returning home after inpatient stroke rehabilitation in Belgium and Switzerland: a multicentric retrospective study
Introduction: Identifying factors associated with discharge destination after inpatient stroke rehabilitation is important for patients and healthcare professionals. It supports discharge planning and prevents delayed discharge.Objective: To identify key variables from socio-demographic and clinical data associated with returning home after inpatient stroke rehabilitation, focusing on patients from three rehabilitation centers in Belgium and Switzerland.Methods: This multicenter retrospective study, conducted in three centers, included 1475 adult patients with stroke admitted to an inpatient rehabilitation unit between December 2012 and June 2021. A logistic regression with backward selection was used to define the model for discharge destination. The dependent variable was the discharge destination (home vs other). The independent variables were selected from the socio-demographic, medical, neurological, care pathway, and functional data and included age, gender, living arrangement, type of stroke, previous stroke, cognitive impairments, independence in grooming, eating, and stair walking.Results: The final model included three variables (independence in stair walking, living arrangement, and cognitive impairment). Stair walking had the strongest association with returning home. Patients who were partially (OR 5.83, 95% CI 3.67-9.26) or fully independent (OR 14.31, 95% CI 9.34-21.93) were more likely to return home than patients who were unable to walk the stairs. The results were similar for subgroups and for discharge and admission data.Conclusion: The study showed that independence in walking stairs is strongly associated with discharge destination. Aligned with another study, these results should be confirmed in further research.
- 2 June 2025
- Published by Taylor & Francis in Neuropsychological Rehabilitation
- Vol. 36 (2) , 442-469
Use of immersive virtual reality to explore visual search behaviour in individuals with visuospatial neglect after stroke
Visuospatial neglect (VSN) is a common post-stroke disorder characterized by impaired attention toward the contralesional hemispace, affecting interaction with the environment. Traditional assessments (e.g., pen-and-paper tests) may not detect subtle deficits in visual search behaviour, particularly in complex, dynamic settings. This study introduces an immersive virtual reality (VR) visual search task designed to assess visual search behaviour across peri- and extra-personal spaces. By integrating eye-and-head tracking, we aimed to characterize search behaviour and evaluate group differences among individuals with stroke with VSN (VSN+, n = 11), those without VSN (VSN-, n = 10), and healthy controls (HC, n = 11). Additionally, we examined how task parameters (target position, spatial context, and difficulty) influenced search behaviour. Results show that VSN+ exhibited least efficient search strategies, with longer search times than VSN- and HC and more extended search paths than HC. Notably, VSN- also demonstrated visual search inefficiencies, suggesting that conventional assessments may overlook subtle visuospatial impairments. Search deficits were most pronounced in extra-personal space and under increased task demands. Despite a limited sample size, these findings highlight the potential of VR-tasks to characterize visual search behaviour. Further research on larger, more diverse samples is needed to evaluate its potential diagnostic utility.
- 2 June 2025
- Published by Taylor & Francis in Neuropsychological Rehabilitation
- Vol. 36 (2) , 442-469
Use of immersive virtual reality to explore visual search behaviour in individuals with visuospatial neglect after stroke
Visuospatial neglect (VSN) is a common post-stroke disorder characterized by impaired attention toward the contralesional hemispace, affecting interaction with the environment. Traditional assessments (e.g., pen-and-paper tests) may not detect subtle deficits in visual search behaviour, particularly in complex, dynamic settings. This study introduces an immersive virtual reality (VR) visual search task designed to assess visual search behaviour across peri- and extra-personal spaces. By integrating eye-and-head tracking, we aimed to characterize search behaviour and evaluate group differences among individuals with stroke with VSN (VSN+, n = 11), those without VSN (VSN-, n = 10), and healthy controls (HC, n = 11). Additionally, we examined how task parameters (target position, spatial context, and difficulty) influenced search behaviour. Results show that VSN+ exhibited least efficient search strategies, with longer search times than VSN- and HC and more extended search paths than HC. Notably, VSN- also demonstrated visual search inefficiencies, suggesting that conventional assessments may overlook subtle visuospatial impairments. Search deficits were most pronounced in extra-personal space and under increased task demands. Despite a limited sample size, these findings highlight the potential of VR-tasks to characterize visual search behaviour. Further research on larger, more diverse samples is needed to evaluate its potential diagnostic utility.
- 3 December 2024
- Published by Elsevier in Brazilian Journal of Physical Therapy
- Vol. 29 (1) , 101153
Association between white matter integrity and lower limb motor impairment after stroke: A systematic review
- 3 December 2024
- Published by Elsevier in Brazilian Journal of Physical Therapy
- Vol. 29 (1) , 101153
Association between white matter integrity and lower limb motor impairment after stroke: A systematic review
- 28 June 2024
- Published by Oxford University Press (OUP) in Archives of Clinical Neuropsychology
- Vol. 39 (8) , 1408-1417
Clinical Assessment of Subjective Visual and Haptic Vertical Norms in Healthy Adults
Accurate verticality perception is essential for daily life activities, such as correctly estimating object orientation in space. This study established normative data for the subjective visual vertical (SVV) and subjective haptic vertical (SHV) using the portable and self-constructable modified Bucket test and Rotating-Column test. Additionally, the contribution of age, sex, and starting position of the line/ column on SVV and SHV accuracy were evaluated. This study, part of the PRECISE project (ClinicalTrials.gov ID NCT05978596), was conducted following the STROBE guidelines. Healthy adults without visual/neurological/vestibular disorders were recruited. Subjective visual vertical and SHV accuracy were described in terms of constant errors (i.e., mean deviation from 0° [true vertical] respecting its direction), unsigned errors (i.e., mean deviation from 0° irrespective of direction), and variability (i.e., intra-individual standard deviation). Sixty participants were evaluated (mean age: 41.14 [SD = 16.74] years). Subjective visual vertical constant errors between −2.82° and 2.90°, unsigned errors up to 2.15°, and variability up to 1.61° are considered normal. Subjective haptic vertical constant errors ranged from −6.94° to 8.18°, unsigned errors up to 6.66° and variability up to 4.25°. Higher ages led to higher SVV unsigned errors and variability. SHV variability was higher in females compared to males. Certain starting positions led to higher SVV and SHV constants and SVV unsigned errors. Normative data are provided for affordable, self-constructable, and portable SVV and SHV tools. These norms are consistent with more sophisticated equipment and can be used to distinguish between normal and abnormal values.
- 28 June 2024
- Published by Oxford University Press (OUP) in Archives of Clinical Neuropsychology
- Vol. 39 (8) , 1408-1417
Clinical Assessment of Subjective Visual and Haptic Vertical Norms in Healthy Adults
Accurate verticality perception is essential for daily life activities, such as correctly estimating object orientation in space. This study established normative data for the subjective visual vertical (SVV) and subjective haptic vertical (SHV) using the portable and self-constructable modified Bucket test and Rotating-Column test. Additionally, the contribution of age, sex, and starting position of the line/ column on SVV and SHV accuracy were evaluated. This study, part of the PRECISE project (ClinicalTrials.gov ID NCT05978596), was conducted following the STROBE guidelines. Healthy adults without visual/neurological/vestibular disorders were recruited. Subjective visual vertical and SHV accuracy were described in terms of constant errors (i.e., mean deviation from 0° [true vertical] respecting its direction), unsigned errors (i.e., mean deviation from 0° irrespective of direction), and variability (i.e., intra-individual standard deviation). Sixty participants were evaluated (mean age: 41.14 [SD = 16.74] years). Subjective visual vertical constant errors between −2.82° and 2.90°, unsigned errors up to 2.15°, and variability up to 1.61° are considered normal. Subjective haptic vertical constant errors ranged from −6.94° to 8.18°, unsigned errors up to 6.66° and variability up to 4.25°. Higher ages led to higher SVV unsigned errors and variability. SHV variability was higher in females compared to males. Certain starting positions led to higher SVV and SHV constants and SVV unsigned errors. Normative data are provided for affordable, self-constructable, and portable SVV and SHV tools. These norms are consistent with more sophisticated equipment and can be used to distinguish between normal and abnormal values.
- 19 August 2023
- Published by SAGE Publications in Neurorehabilitation and Neural Repair
- Vol. 37 (8) , 530-544
Recovery of Quiet Standing Balance and Lower Limb Motor Impairment Early Poststroke: How Are They Related?
Background: Recovery of quiet standing balance early poststroke has been poorly investigated using repeated measurements. Objective: To investigate (1) the time course of steady-state balance in terms of postural stability and inter-limb symmetry, and (2) longitudinal associations with lower limb motor recovery in the first 3 months poststroke. Methods: Forty-eight hemiparetic subjects (age: 58.9 ± 16.1 years) were evaluated at weeks 3, 5, 8, and 12 poststroke. Motor impairments concerned the Fugl-Meyer assessment (FM-LE) and Motricity Index total score (MI-LE) or ankle item separately (MI-ankle). Postural stability during quiet 2-legged stance was calculated as the net center-of-pressure area (COPArea) and direction-dependent velocities (COPVel-ML and COPVel-AP). Dynamic control asymmetry (DCA) and weight-bearing asymmetry (WBA) estimated inter-limb symmetries in balance control and loading. Linear mixed models determined (1) time-dependent change and (2) the between- and within-subject associations between motor impairments and postural stability or inter-limb symmetry. Results: Time-dependent improvements were significant for FM-LE, MI-LE, MI-ankle, COPArea, COPVel-ML, and COPVel-AP, and tended to plateau by week 8. DCA and WBA did not exhibit significant change. Between-subject analyses yielded significant regression coefficients for FM-LE, MI-LE, and MI-ankle scores with COPArea, COPVel-ML, and COPVel-AP up until week 8, and with WBA until week 12. Within-subject regression coefficients of motor recovery with change in COPArea, COPVel-ML, COPVel-AP, DCA, or WBA were generally non-significant. Conclusions: Postural stability improved significantly in the first 8 weeks poststroke, independent of lower limb motor recovery at the most affected side within subjects. Our findings suggest that subjects preferred to compensate with their less affected side, making metrics reflecting inter-limb asymmetries in balance invariant for change early poststroke. Clinical Trial Registration: Clinicaltrials.gov. unique identifier NCT03728036.
- 19 August 2023
- Published by SAGE Publications in Neurorehabilitation and Neural Repair
- Vol. 37 (8) , 530-544
Recovery of Quiet Standing Balance and Lower Limb Motor Impairment Early Poststroke: How Are They Related?
Background: Recovery of quiet standing balance early poststroke has been poorly investigated using repeated measurements. Objective: To investigate (1) the time course of steady-state balance in terms of postural stability and inter-limb symmetry, and (2) longitudinal associations with lower limb motor recovery in the first 3 months poststroke. Methods: Forty-eight hemiparetic subjects (age: 58.9 ± 16.1 years) were evaluated at weeks 3, 5, 8, and 12 poststroke. Motor impairments concerned the Fugl-Meyer assessment (FM-LE) and Motricity Index total score (MI-LE) or ankle item separately (MI-ankle). Postural stability during quiet 2-legged stance was calculated as the net center-of-pressure area (COPArea) and direction-dependent velocities (COPVel-ML and COPVel-AP). Dynamic control asymmetry (DCA) and weight-bearing asymmetry (WBA) estimated inter-limb symmetries in balance control and loading. Linear mixed models determined (1) time-dependent change and (2) the between- and within-subject associations between motor impairments and postural stability or inter-limb symmetry. Results: Time-dependent improvements were significant for FM-LE, MI-LE, MI-ankle, COPArea, COPVel-ML, and COPVel-AP, and tended to plateau by week 8. DCA and WBA did not exhibit significant change. Between-subject analyses yielded significant regression coefficients for FM-LE, MI-LE, and MI-ankle scores with COPArea, COPVel-ML, and COPVel-AP up until week 8, and with WBA until week 12. Within-subject regression coefficients of motor recovery with change in COPArea, COPVel-ML, COPVel-AP, DCA, or WBA were generally non-significant. Conclusions: Postural stability improved significantly in the first 8 weeks poststroke, independent of lower limb motor recovery at the most affected side within subjects. Our findings suggest that subjects preferred to compensate with their less affected side, making metrics reflecting inter-limb asymmetries in balance invariant for change early poststroke. Clinical Trial Registration: Clinicaltrials.gov. unique identifier NCT03728036.
- Open Access
Gait Stride Length Estimation Using Embedded Machine Learning
Introduction. Spatiotemporal gait parameters, e.g., gait stride length, are measurements that are classically derived from instrumented gait analysis. Today, different solutions are available for gait assessment outside the laboratory, specifically for spatiotemporal gait parameters. Such solutions are wearable devices that comprise an inertial measurement unit (IMU) sensor and a microcontroller (MCU). However, these existing wearable devices are resource-constrained. They contain a processing unit with limited processing and memory capabilities which limit the use of machine learning to estimate spatiotemporal gait parameters directly on the device. The solution for this limitation is embedded machine learning or tiny machine learning (tinyML). This study aims to create a machine-learning model for gait stride length estimation deployable on a microcontroller. Materials and Method. Starting from a dataset consisting of 4467 gait strides from 15 healthy people, measured by IMU sensor, and using state-of-the-art machine learning frameworks and machine learning operations (MLOps) tools, a multilayer 1D convolutional float32 and int8 model for gait stride length estimation was developed. Results. The developed float32 model demonstrated a mean accuracy and precision of 0.23 ± 4.3 cm, and the int8 model demonstrated a mean accuracy and precision of 0.07 ± 4.3 cm. The memory usage for the float32 model was 284.5 kB flash and 31.9 kB RAM. The int8 model memory usage was 91.6 kB flash and 13.6 kB RAM. Both models were able to be deployed on a Cortex-M4F 64 MHz microcontroller with 1 MB flash memory and 256 kB RAM. Conclusions. This study shows that estimating gait stride length directly on a microcontroller is feasible and demonstrates the potential of embedded machine learning, or tinyML, in designing wearable sensor devices for gait analysis.
Open AccessGait Stride Length Estimation Using Embedded Machine Learning
Introduction. Spatiotemporal gait parameters, e.g., gait stride length, are measurements that are classically derived from instrumented gait analysis. Today, different solutions are available for gait assessment outside the laboratory, specifically for spatiotemporal gait parameters. Such solutions are wearable devices that comprise an inertial measurement unit (IMU) sensor and a microcontroller (MCU). However, these existing wearable devices are resource-constrained. They contain a processing unit with limited processing and memory capabilities which limit the use of machine learning to estimate spatiotemporal gait parameters directly on the device. The solution for this limitation is embedded machine learning or tiny machine learning (tinyML). This study aims to create a machine-learning model for gait stride length estimation deployable on a microcontroller. Materials and Method. Starting from a dataset consisting of 4467 gait strides from 15 healthy people, measured by IMU sensor, and using state-of-the-art machine learning frameworks and machine learning operations (MLOps) tools, a multilayer 1D convolutional float32 and int8 model for gait stride length estimation was developed. Results. The developed float32 model demonstrated a mean accuracy and precision of 0.23 ± 4.3 cm, and the int8 model demonstrated a mean accuracy and precision of 0.07 ± 4.3 cm. The memory usage for the float32 model was 284.5 kB flash and 31.9 kB RAM. The int8 model memory usage was 91.6 kB flash and 13.6 kB RAM. Both models were able to be deployed on a Cortex-M4F 64 MHz microcontroller with 1 MB flash memory and 256 kB RAM. Conclusions. This study shows that estimating gait stride length directly on a microcontroller is feasible and demonstrates the potential of embedded machine learning, or tinyML, in designing wearable sensor devices for gait analysis.
- 19 July 2023
- Published by Springer Nature
- p. 98-105
Applications of Using 4D Scanning Technologies in Biomechanics
- 19 July 2023
- Published by Springer Nature
- p. 98-105
Applications of Using 4D Scanning Technologies in Biomechanics